
Language Wars Truce Accepted
(with conditions)
Peter Tiersma
I
N A RECENT ISSUE of this journal, Bryan ment over earlier versions of this venerable
Garner, who is becoming increasingly dictionary. He has also written several books
prominent as a legal lexicographer, ad- on legal usage and style, and most recently a
dressed what he describes as a “war” between manual on usage in general, Garner’s Modern
descriptive and prescriptive grammarians. American Usage.
Although the distinction between preAs the title of his article suggests, he proposes that we should be Making Peace in the scriptive and descriptive practices might
Language Wars.¹ After describing himself as seem rather arcane, Garner’s influence in the
a “descriptive prescriber” who nonetheless world of legal lexicography makes his deplaces himself “mostly in the prescriptive fense of prescriptivism, nuanced as it is, pocamp,”² Garner proposes a truce. The terms tentially worrisome. It matters whether he is
are essentially that the describers should let a descriptivist, a prescriptivist, or some kind
the prescribers prescribe, and vice versa. In of hybrid. This is especially true because he
other words, live and let live. It is an appeal to is not just a style guru, but also a lexicograpeace and tolerance, which should resonate pher in his role as editor of Black’s, as well as
an expert on usage, as evidenced by his Dicwell with modern readers.
As the self-appointed negotiator for the tionary of Modern Legal Usage. While style
field of linguistics, I am prepared to accept gurus can, almost by definition, be expected
the truce that Garner proposes, subject to to display a certain measure of prescriptivism, anything called a “dictionary” should be
certain conditions and clarifications.
Many of us know who Bryan Garner is. based entirely on careful observation and deHe is the editor in chief of Black’s Law Dic- scription of actual usage.
tionary, recently released in its eighth edition.
At the least, those who rely on Garner’s
His editorship has produced a vast improve- work, which includes much of the legal proPeter Tiersma is a professor of law and Joseph Scott Fellow at Loyola Law School, Los Angeles.
 Bryan A. Garner, Making Peace in the Language Wars, 7 Green Bag 2d 227 (2004).
2 Id. at 230.
Pe t e r Ti e r s m a
fession, should know what they’re buying that it is a rule of English that sentences
into. I doubt that purchasers of Garner’s ex- should not end in a preposition. Other prepanding oeuvre will want their money back scriptive rules include the maxim that ain’t
after they understand the lexicographer’s role is not a word, that you should never use but
more clearly. But hopefully³ they will have a or however at the beginning of a sentence, or
better appreciation of what it means to “look that the relative pronoun which should only
up” a word in the dictionary or to consult introduce a nonrestrictive relative clause.
The prescriptivist notion of a rule is therefore
one of his usage or style manuals.
My conclusion will be that Garner is more very similar to the legal conception. A rule is
of a descriptivist than he admits, and less of a norm that governs behavior. Its purpose is
a prescriptivist than he claims. And, from a not to describe how people act, but rather to
linguistic point of view, that’s a complement. modify how they act.
In a legal context it is generally clear who
Or is it compliment?
has the authority to make rules. The ConDescriptivism vs. Prescriptivism stitution gives Congress lawmaking power.
This raises an interesting question: if rules
When linguists distinguish between descrip- relating to language can be prescriptive – if
tive and prescriptive, it is usually with respect they can dictate how we ought to speak and
to rules of grammar or usage. Much of the write – who decides what those rules are or
controversy in some way or other relates to should be? Where do prescribers derive the
the nature of a rule in the context of language, power to legislate how the rest of us ought to
an issue that lawyers should be able to relate use language?
to. An example may be found in the previous
A linguist (the adjective “descriptive” is
sentence. Many people claim that there is a virtually redundant in this context) has a
rule in English that a sentence should never far different conception of rules. The rules
end with a preposition. Hence, the “correct” that linguists posit are meant to describe or
way to phrase the clause in question would explain regularities that naturally occur in
be “an issue to which lawyers should be able language and that are passed on from one
to relate.”
generation to the next with little conscious
Of course, stranded prepositions are ex- thought. If a rule posited by a particular lintremely common in ordinary speech. Most guist is not an accurate portrayal of actual
of us recall the comment attributed to Win- linguistic behavior, it must be modified to
ston Churchill: “This is the sort of English correctly reflect the data. The linguistic noup with which I will not put.”⁴ Churchill, or tion of a rule is therefore similar to how the
whoever else might have said it, was clearly concept is used in the sciences. It is based on
spoofing this purported rule.
hard data: how people actually use language
A prescriber thus uses the word rule in a to communicate.
normative sense. The rule prescribes or dicAn important feature of rules, as describtates how a person ought to speak. As we ers understand the concept, is that they are
have seen, there are prescriptivists who claim passed on naturally from one generation
3
I was tempted to check what Garner might have to say about the use of “hopefully” but decided that I
should rely on my own instinct, which is that its use in the meaning “I hope” is sufficiently well established to allow the word to appear in this sense in relatively formal prose.
4 John Bartlett, Familiar Quotations 746 (5th ed. 980).
2 82
8 G R E E N B A G 2 D 2 81
L a n g u a g e Wa r s Tr u c e A c c e p t e d (w i t h c o n d i t i o n s)
to the next. People sometimes consciously out that there is a critical difference between
learn language, of course, and parents may a speaker’s competence and performance:
sometimes try to instruct their offspring
Linguistic theory is concerned primarily
on certain aspects of the language. But conwith an ideal speaker-listener, in a comscious instruction and learning are not espletely homogeneous speech-community, who knows its language perfectly
sential. Barring a serious mental or physical
and is unaffected by such grammatically
impediment, every human being who has
irrelevant conditions as memory limitasufficient exposure to a language as a child
tions,
distractions, shifts of attention and
learns to speak it. Prescriptive rules, on the
interest, and errors (random or characother hand, must be explicitly taught and
teristic) in applying his knowledge of
learned.
the language in actual performance.⁷
Incidentally, the notion of a descriptive
rule is not entirely foreign to the law, even Although virtually all of Chomsky’s ideas
if most legal rules are prescriptive. A good have been the subject of debate and revision,
example of descriptive rules are those in the most linguists would agree that we need
various restatements. The rules contained to distinguish between the sentences that
in the restatements are meant to restate or speakers produce in the real world (perfordescribe the general common law as devel- mance) and the sentences that their internal
oped in the United States.⁵ They are not grammars would produce under ideal cirpurely descriptive, of course, because the cumstances (competence). In other words,
temptation to improve or clarify is likely to we all make mistakes in actual speech probe irresistible. But the main focus of the en- duction. Unless we ignore random errors, it
terprise is descriptive. If for no other reason, would be impossible to posit general rules
the restatements are descriptive because the of a language.
lawyers and judges who draft them have no
Another area of difference between deauthority to prescribe anything.
scribers and prescribers is what they mean
There is a common misconception that by “usage.” Both groups agree that usage
because descriptive rules are based on actual – what people actually say and write – is at
behavior, linguists believe that a speaker, by the heart of the matter. For linguists, any dedefinition, cannot make a mistake. Anything scription of a language must correspond to
goes. Garner seems to belong to this camp actual usage. Garner also acknowledges the
when he criticizes describers as believing importance of usage. He carefully monitors
“that native speakers of English can’t make a usage by the legal community, using it to inmistake and that usage guides are therefore form his stylistic and lexical judgments. And
superfluous.”⁶ It may be that some linguists, he admits that if a prescriptive rule of lanfilled with zeal to promote a more scientific guage no longer corresponds to at least some
approach to language, have made rash state- usage in the speech community, it is time to
ments of the kind that Garner suggests. It is toss in the towel.
not the standard view in the field, however.
With respect to usage, the main difference
No less a figure than Noam Chomsky, one between describers and prescribers is that
of the founding fathers of linguistics, pointed the latter focus on usage by those they deem
5 See Harvey S. Perlman, The Restatement Process, 0 Kan. J. L.  Pub. Pol’y 2 (2000).
6 Garner, Green Bag, at 228.
7 Noam Chomsky, Aspects of the Theory of Syntax 3 (965).
Articles  Spring 2005
2 83
Pe t e r Ti e r s m a
Samuel Johnson’s prescription or description of “usage” in the fifth edition of his Dictionary.
the “better” users of language. They also tend
to focus more on written language. Linguists,
in contrast, are interested in the usage of the
entire speech community.
Of course, since linguists are describers,
they are aware of the existence of “proper”
English. Generally, they prefer to call it something like “Standard American English” (or
SAE). The origins of SAE must be sought
in England, where a type of standard written language emerged several centuries ago,
based largely on the dialects of London and
8
the East Midlands.⁸ SAE is to some extent
the daughter of standard British English, but
at the same time it reflects the influences of
the many regional varieties of English spoken throughout the United States.
Linguists often say, somewhat disparagingly, that SAE is just another dialect of
English. There is a sense in which this is true,
as the origins of SAE indicate. The point is
that there is no inherent reason to privilege
SAE. Why should the speech of New Yorkers or African Americans be considered less
Barbara A. Fennell, A History of English: A Sociolinguistic Approach 22–3 (200).
2 84
8 G R E E N B A G 2 D 2 81
L a n g u a g e Wa r s Tr u c e A c c e p t e d (w i t h c o n d i t i o n s)
correct than the speech of upper-class white report at work, it may not be the best choice
Midwesterners, whose native dialect is quite when hanging out with your friends in front
similar to SAE? Those who grow up speak- of the liquor store.
On another level, however, the battle is
ing dialects of English that strongly diverge
from SAE are clearly at a disadvantage com- more than semantic. The difference between
pared to children growing up in homes where a descriptive and prescriptive approach has
a dialect close to SAE is spoken.
important real-world implications with reBut if all we can say about SAE is that spect to the art or science of compiling dicit is another dialect of English, we have se- tionaries and, to a lesser extent, style manuals.
verely mischaracterized the role of standard We have now come back to the original point
English in our society. It would simply be of this essay: it matters whether Garner is a
a bad description of reality. The statement describer or a prescriber, because he has becan be justified to some extent on linguis- come a force to be reckoned with in both the
tic grounds, but only at the cost of ignoring field of legal lexicography and what we might
important social attitudes towards language. call the “style business.”
Linguists tend not to like such attitudes, but
we cannot ignore their existence.
Why Garner Should
Thus, linguists must ultimately recognize
Be a Describer
(as most do) that speakers of English realize
that there is such a thing as standard Eng- One of Garner’s roles is the editorship of
lish. They also recognize that many speakers Black’s Law Dictionary. From what I can tell,
aspire, for a variety of social and economic he is doing a fine job. Black’s has improved
reasons, to speak and write standard English. because Garner has instituted sound lexiBecause few people acquire this variety of cographic principles. Most importantly, his
English naturally, they need to learn it. This, definitions are based upon actual usage by
of course, is where Bryan Garner and other the profession. He and his staff have combed
prescriptivists enter the picture.
through a vast assortment of legal texts, inTo some extent, therefore, the battle be- cluding not just the predictable statutes and
tween describers and prescribers can be said judicial opinions, but also legal encyclopedias,
to boil down to semantics. Linguists will restatements, law reviews, treatises, and even
always complain when prescribers proclaim the Nutshell series.⁹ Garner also had three
that they are teaching people to speak “cor- distinguished scholars vet the entire manurectly.” But if the prescribers would suggest script of the seventh edition, and asked an
that their mission is to help interested peo- additional 30 lawyers, judges, and academics
ple learn how to speak and write according to to read parts of it.¹⁰ The recently-published
the norms of standard English, I imagine that eighth edition received input from even more
few linguists would seriously object. This is experts, and the entire dictionary was reespecially true if prescribers would also ad- viewed by a panel of academic contributors.¹¹
mit that there are multiple varieties of Eng- Specialists will probably quibble about some
lish, and that while standard English may be details, but all in all it is a remarkable revitalthe most appropriate option when writing a ization of an old workhorse.
9
0

See Black’s Law Dictionary (7th ed. 999, appendix G).
Id. at xiii.
Black’s Law Dictionary v (8th ed. 2004). I was one of those contributors.
Articles  Spring 2005
285
Pe t e r Ti e r s m a
Garner’s reworking of Black’s illustrates ary of Modern English Usage (second edition),
not merely that he is aware of modern lexi- which goes so far as to express a preference
cographic principles, but also that – contrary for hypo.
Of course, part of the knowledge that
to his stated views in this journal – he is very
speakers have about words such as some of
much a describer.
Lexicographers of the past sometimes the above is that they are considered inforlet their own predispositions influence how mal, jargon, obscene, and so forth. Speakers’
they defined a word. They might leave out attitudes about these words are very real.
a word entirely, or ignore a common mean- Usage tags thus have an important funcing of a word, because they deemed it unac- tion. But even usage tags should be based
ceptable. Today, it seems to me, most people on sound descriptive practices, not prescripwould agree that a dictionary should not tion. A term should be marked slang because
engage in self-censorship, as the Academie speakers generally regard it as such, not
Française does for its dictionary of French. If merely because the editors do so. There will
you hear people talking about spamming or be difficult judgment calls, of course. When
blogging or how many megs of ram their com- exactly a word like spam (in the sense of unputer has, shouldn’t you be able to find out in wanted email) ceases to be slang cannot be
a dictionary what speakers mean when they precisely delineated. But if legislative boduse these words? And even though school ies are considering anti-spam legislation, it is
teachers may tell you that ain’t is not a word, hard to deny that the word has entered the
shouldn’t an immigrant learning English or general vocabulary.
a foreign high-school student reading HuckThus, when Garner, the editor-in-chief
leberry Finn be able to find its meaning in a of Black’s, refers to himself as primarily a
dictionary? You can’t function very well as a prescriber, it makes me somewhat nervous. I
speaker of English without such knowledge. am only slightly nervous, because in actual
If a dictionary is a repository of information practice he seems to be a sound describer.
about words, it needs to include items that The danger is that in the act of describing
we might consider slang, neologisms, dialect, he may unwittingly slip into prescription.
and any other word that someone might There is a fundamental distinction between
want to look up.
is and ought, and only the former belongs in
Likewise, the definitions should be based a dictionary.
on actual usage, not on someone’s idea of
how a word ought to be used. I once had a
Is There Still a Role
legal secretary who sent back my memos
for Precribers?
with comments such as “conclusory is not an
English word” and “hypothetical should not Garner has also gained increasing promibe used as a noun.” Interestingly, the seventh nence as what we might call a style or usage
edition of Black’s provides a definition of guru. Style gurus are traditionally allowed –
conclusory (although my spell-checker keeps even expected – to be moderately prescripobjecting as I write this). Oddly, the seventh tive. After all, people look to them for adedition does not explain what a hypothetical vice. They want to know how to speak and
is, despite its ubiquitous use by lawyers and write properly, which inherently involves a
law professors. But the word appears with- certain amount of judgment. When people
out pejorative comment in Garner’s Diction- consult Miss Manners for advice on how
286
8 G R E E N B A G 2 D 2 81
L a n g u a g e Wa r s Tr u c e A c c e p t e d (w i t h c o n d i t i o n s)
to arrange the silverware at a formal dinner, they want a clear-cut answer. The same
goes for questions that people have about
choosing and arranging words. Is it that or
which, who or whom, complement or compliment? Does someone graduate college or
graduate from college? The data prove or the
data proves?
Comparing prescribers to Miss Manners may seem to trivialize the useful work
that Garner and other prescribers have done
and the scholarship that underlies their efforts. That is certainly not my intention. Yet
it is worth noting some interesting overlap. A
critical question that we could ask of both
Mr. Style and Miss Manners is how they
know what is “proper” or “correct.” Miss Manners would probably reply that she learned
how to behave properly by observing people
who have good manners. Of course, what it
means to have good manners, and who has
them, is debatable. Certainly in the past,
however, there was a group of people – often wealthy white Protestants – who were
viewed as knowing how to act properly, and
many people strove (or is it strived?) to imitate them.
So, how do style and usage gurus know
what to prescribe? Like Miss Manners, they
mostly base their judgments on the linguistic habits of people who they believe know
how to how to speak and write “properly.”
Prescribers may believe that they intuitively
have a good sense of what is correct and
what is not, but such a sense surely derives
from a great deal of reading works that one
considers to be good English and unconsciously absorbing the rules that the writers
of those works follow. More careful prescribers, like Garner, go a step further by explicitly
researching the habits of good writers to justify their pronouncements.
Relying on the usage of good writers raises an obvious question: who decides which
writers are the models to emulate? In Great
Britain, people once aspired to speak the
King’s English, since his English was proper
by definition. So, should we Americans try
to speak like President Bush? If the issue is
legal usage, do we imitate Holmes? Cardozo?
O’Connor? Scalia? Richard Posner, Duncan
Kennedy, or whoever else happens to be our
idol?
I suppose that the difficulties of making
such decisions is why we appreciate the work
of prescribers. We trust them to figure out
who the good writers are, to determine what
those writers would do when confronting a
thorny lexical dilemma or syntactic quandary,
and to package the answer to our question
into a succinct paragraph that provides us
with a clear-cut answer.
Thus, what it boils down to is this: can
we trust Brian Garner and other prescribers to decide for us what is correct English
and what is not? Noah Webster spoke out
against “literary governors” who “dictate to a
nation the rules of speaking, with the same
imperiousness as a tyrant gives orders to his
vassals.”¹² Ironically, Webster asked Justice
John Marshall to endorse his dictionary of
American English. Marshall refused, stating
that in a democracy no one can dictate usage.¹³ Maybe so, but that doesn’t mean people
can’t seek some good advice now and then.
Overall, Garner’s work as usage or style
guru strikes a reasonable balance between
adhering to rules of the language that are
fairly well established, and letting go when
the tides of change are simply too strong.
2
Quoted in Naomi S. Baron, Alphabet to Email: How Written English Evolved and Where It’s Heading 34
(2000).
3 Shirley Brice Heath, A National Language Academy?: Debate in the New Nation, 89 Linguistics 9, 33
(977).
Articles  Spring 2005
2 87
Pe t e r Ti e r s m a
Thus, his Dictionary of Legal Usage¹⁴ refers to
the rule about not ending a sentence with a
preposition as “spurious” (p. 686). He allows
the splitting of infinitives, albeit cautiously,
quoting with apparent approval one of the
opening lines in the Star Trek television series (“to boldly go where no man has gone
before”) (p. 823). He roundly condemns the
rule that prohibits starting a sentence with
and or or, referring to it as “rank superstition”
(p. 55). And he accepts (or declines to condemn) useful innovations, including legal neologisms like hedonic damages or palimony.
Whatever some of my former colleagues
from the world of linguistics might say, just
about anyone can use a good usage dictionary or style manual now and then. Even
though SAE is to some extent artificial, and
even though in a democracy no one can tell
us how we ought to speak, it is undeniable
that certain conventions regarding standard English – especially the written variety
– have become well established. Any writer
who flouts those conventions has less credibility with readers who are familiar with the
conventions and expect them to be followed.
Yet at times Garner drifts too far from
actual usage and is therefore overly prescriptive, at least for my taste. Let us return to
the past tense of strive. The second edition
of his Dictionary of Modern Legal Usage suggests that strived is incorrect and should be
replaced by strove. That also reflects my own
dialect of English. Yet my copy of the American Heritage Dictionary also lists strived as
a possibility (although it is tagged as “rare”).¹⁵
Moreover, a search of the internet using the
search engine Google found strived approximately 47,000 times, while the results for
strove were 250,000. Strove is indeed used
4
5
6
7
288
more often, but the difference is not enormous, and it hardly merits tagging strived as
“rare” or condemning it outright.
There is actually a lot of variation in the
past tense and past participle forms of verbs.
The problem that I (and most linguists) have
with typical prescribers is that instead of acknowledging the existence of variation, they
seem to have an almost irresistible urge to
label one of the two variants as wrong or improper.
Another verb, prove, is an interesting illustration. Here, the contentious issue is the
past participle, which can be either proved or
proven (as in “the government has not proved/
proven its case.”) Garner takes a strong position in favor of proved: “Proved is the universally preferred past tense of prove; proven
… properly exists only as an adjective.”¹⁶ Of
course, if Garner is correct, the phrase “innocent until proven guilty” is ungrammatical.
The famous Scottish verdict should be “not
proved” in place of “not proven.”
It turns out that proved is indeed more
common as a past participle among “better” writers than is proven. The American
Heritage Dictionary has taken an innovative
approach to such controversial usage issues
by assembling a usage panel consisting of
dozens of distinguished writers. Only about
one quarter of the panel preferred proven.¹⁷
The U.S. Supreme Court also seems to have
a preference for has proved (occurring 25
times in the sct database on Westlaw), as opposed to has proven (occurring 5 times).
Nonetheless, I’m sticking to proven. Preference for proved is hardly universal, even
among good writers, and I see no reason to
conform my language whenever I happen
to be in the minority. As a matter of fact, I
Second edition, 995.
The reference is to the first edition.
Dictionary of Modern Legal Usage 445 (987).
Reference is to the first edition.
8 G R E E N B A G 2 D 2 81
L a n g u a g e Wa r s Tr u c e A c c e p t e d (w i t h c o n d i t i o n s)
Conditions for
may well be in the majority with respect to
actual usage. A search on the internet for the
Accepting Truce
phrase has proved, once again using Google,
found around ,080,000 occurrences. Has As a member in good standing of the Linguisproven, on the other hand, produced around tic Society of America, I am now prepared
,600,000 hits. The internet is an unruly to accept Garner’s proposed truce, subject to
place sometimes, so these results are only some important conditions.
. Any work labeled “dictionary,” including
suggestive, but it may well be that proven is
Black’s, should be purely descriptive. I use the
actually more common than proved.
My point is not that Garner is dispens- word descriptive here in a relatively broad
ing bad advice here, even if it goes against sense, encompassing not just actual usage in
my own personal preference. Rather, some- the legal community, but also (with respect
what less prescription and somewhat more to usage tags) the attitudes that the profesdescription would be nice. Admittedly, there sion has towards language. Thus, if a term is
are a lot of people who seem to be hankering in common use among lawyers, but mostly in
for a simple thumbs-up or thumbs-down on spoken and informal contexts, it is approprisome of these issues. These are the sort of ate to label it as slang. In fact, it seems to me
folks who are distressed by the fact that there that any definition of words like punies or inare no less than three adjectives referring to cidentals would be incomplete without addScotland (Scots, Scottish, and Scotch) and who ing such information, because the fact that
are desperate for someone to tell them which these words are slang is part of the knowlone is “correct.”
edge that speakers have about them.
At the same time, I imagine that many
As far as I can tell, Garner already strives
more educated writers, such as lawyers and to base the definitions in Black’s on actual usjudges, would prefer a realistic description age, so he should not find it difficult to accept
of actual usage, so they can evaluate the this first condition. The reason that I include
evidence for themselves and make their own it is that in some sense there is a conflict of
decisions. In the case of prove, an accurate interest between style gurus, who are inclined
statement would be that both proved and to condemn certain usages and beatify othproven are widely used as past participles ers, and lexicographers, who should faithfully
in speech, and that both occur in writing, record any usage that is likely to be encounalthough proved seems to predominate in tered by those who consult a dictionary. Thus,
more formal written contexts. This gives me when he wears his lexicographer’s hat, Garner
the sort of information I need to decide the should be a describer, pure and simple.
issue for myself.
2. Prescription should be based on sound deGarner is free to argue in favor of proved scriptive observations. Any pronouncements
or strove. He can even prescribe it to those of what is “proper” or “correct” should be
who seek his advice. He should do so, how- grounded in the actual usage of well-regardever, in light of solid descriptive evidence. ed legal writers. Like all of us, Garner is enThe users of his works deserve to know when titled to his pet peeves, but when something
something really is a rule of standard Eng- is mainly his own preference, or where he belish, accepted by virtually all good writers, lieves that current usage should be changed,
and when there is variation even among the he should fully disclose that there are fine
best literary or judicial talent.
writers who do not agree.
Articles  Spring 2005
289
Pe t e r Ti e r s m a
Although Garner has strong opinions even in his role as usage or style guru, a careful
about some issues, he seems to be open to observer of actual usage. He is, in other words,
persuasion by evidence of actual usage. For primarily a describer. What he describes is not
instance, in the first edition of his Diction- the English that is spoken by ordinary people,
ary of Legal Style, which I skimmed through nor even his own dialect, but the English that
when I was preparing my book on legal lan- is written by good legal writers. He might as
guage,¹⁸ I was stunned to see that he consid- well admit it, because basing his advice on the
ered chid to be the preferred past tense and usage of good writers can only enhance the
past participle form of chide in the United authoritativeness of his work.
States (p. ). I can’t recall ever hearing the
form. Apparently, I was not the only person

with this reaction; the second edition states
that chided is preferred.¹⁹
Anyway, Bryan, I am ready to sign the truce,
I therefore challenge Garner to come out subject to these modest conditions. Or mayof the closet and openly declare that he is, be you’d rather just defect?
8 Peter M. Tiersma, Legal Language (999).
9 In any event, it seems an odd item to add to a dictionary of legal usage.
290
8 G R E E N B A G 2 D 2 81